Materials Science and Engineering C 47 (2015) 123134

Contents lists available at ScienceDirect

Materials Science and Engineering C

journal homepage: www.elsevier.com/locate/msec

Mechanical and biological properties of oxidized horn keratin

Quanbin Zhang a, Guanghua Shan b, Ping Cao c, Jia He c, Zhongshi Lin c, Yaoxiong Huang a, Ningjian Ao a,
a
b
c

Department of Biomedical Engineering, Jinan University, Guangzhou 510632, China

Cardiology, The First Afliated Hospital of Jinan University, Guangzhou 510632, China
Shenzhen Testing Center of Medical Devices, Shenzhen 518057, China

a r t i c l e

i n f o

Article history:
Received 21 September 2014
Accepted 12 November 2014
Available online 14 November 2014
Keywords:
Horn keratin
Protein oxidation
Mechanical properties
Biocompatibility
Biomaterial

a b s t r a c t
The goal of this study was to investigate the mechanical and biological properties of oxidized keratin materials,
which were obtained by using buffalo horns to oxidize. It could provide a way to evaluate their potential for clinical translatability. The characterization on their composition, mechanical properties, and biological responses
was performed. It is found that the oxidation process could lead the disulde bond to break down and then to
form sulfonic acid, or even make partial peptide chain to be fragment for the new modication of amino acid.
Hence the oxidized horn keratins have lower thermal stability and hydrolytic stability in comparison with
horn keratin, but the degradation products of oxidized horn keratins have no signicant difference. In addition,
the mechanical properties of oxidized horn keratins are poorer than that of horn keratin, but the oxidized horn
keratins still have disulde bonds to form a three-dimensional structure, which benets for their mechanical
properties. The fracture toughness of oxidized horn keratins increases with the increase in the degree of oxidation. After oxidation, the oxidized horn keratins have lower cytotoxicity and lower hemolysis ratio. Moreover,
when the oxidized horn keratins, as well as different concentration of degradation products of oxidized horn keratins, are directly in contact with platelet-rich plasma, platelets are not activated. It suggests that the oxidized
horn keratins have good hemocompatibility, without triggering blood thrombosis. The implantation experiment
in vivo also demonstrates that the oxidized horn keratins are compatible with the tissue, because there are minimal brous capsule and less of inltration of host cells, without causing serious inammation. In summary, the
oxidized horn keratins can act as implanted biomaterial devices that are directly in contact with blood and tissue.
2014 Elsevier B.V. All rights reserved.

1. Introduction
Keratin is one of the most abundant proteins [1]. The rst study of
keratin used as biomaterial was published by Noishiki and his colleagues. They coated a heparinized keratin derivative onto a vascular
graft, which was implanted into a dog, without thrombosis for more
than 200 days [2]. In fact, keratin is the major component of any tissue
of a living organism, such as hair, wool, feathers and horns, which is
often associated with various biological functions. For example, they
can serve as a barrier for environmental stress, regulate moisture and
communicate with others. Keratin can be distinguished as soft and
hard keratin [3,4]. Soft keratin is often found in epidermis and calluses
and it has lower sulfur content. However, hard keratin has higher sulfur
content. Hard keratin can be classied into two groups. The one is hard
-keratin, which is found in mammalian epidermal appendages, such as
horns, hairs and nails, and the other one is -keratin, which is found in
avian and reptilian tissues. The -keratin has an -helical coil structure,
but the -keratin has a twisted -sheet structure. In recent years, the

Corresponding author.
E-mail address: [email protected] (N. Ao).

http://dx.doi.org/10.1016/j.msec.2014.11.051
0928-4931/ 2014 Elsevier B.V. All rights reserved.

keratinous materials have attracted increasing attention, such as equine

hoof [5], bovid hoof [6], wool [7], and especially the sheep horn [8].
Horn appears on animals coming from the bovid family, which includes cattle, sheep, and waterbuck, and it is composed of a keratinous
sheath overlying a bony core [9]. The horn keratin is a hard -keratin,
and has crystalline ber phase and amorphous matrix phase [10]. The
crystalline phase contains microbrils with -helical structure, but the
amorphous phase is made up of microbrils with non-helical structure
and other morphological components. In a horn, the keratin bers are
substantially parallel to the growth direction and are stacked with
each other to form lamellar structure [11]. Besides, keratin bers are
embedded in an amorphous and non-brous protein matrix through
many bonds, such as disulde bonds, hydrogen bonds, van der Waals
forces and ionic interaction [10]. Thus, the horn keratin forms an excellent biological model with a hierarchical structure from nanometer to
micrometer scale [12]. As a result, horn keratin has good performances,
such as high toughness, stiffness and strength. However, the stable
three-dimensional structure of horn, which is formed by disulde bridges and other crosslinks, makes keratin have high chemical stability in
physiological environment, resulting in its non-degradability, which is
hindering its application [13]. Many extraction methods for the soluble
keratin have been studied extensively, such as oxidation and reduction

124

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

[1417]. The soluble keratin would have advantages in several elds,

such as wound care, tissue reconstruction, cell seeding and diffusion,
and drug delivery [18]. But it doesn't have three-dimensional structure,
which makes it have poor mechanical property and processability. Thus,
the practical applications of soluble keratin are restricted [19]. Therefore, it is desired to prepare a keratin with both good mechanical property and degradation.
Therefore, in order to design a sort of degradable keratin with threedimensional structure, we prepared an oxidized horn keratin biomaterial
by using oxidative conditions and investigated its general chemical, mechanical, and biological properties. In addition, the hemocompatibility of
the oxidized horn keratin was studied by, hemolysis ratio, platelet adhesion tests, partial thromboplastin time (PTT), activated partial thromboplastin time (APTT) and brinogen (Fib) assay.
2. Materials and methods
2.1. Sample preparation
The buffalo horns from subadult and healthy bovine were obtained
within 24 h after slaughter from a local slaughterhouse (butchered for
dietary reasons; Baiyun District, Guangzhou, China). The horn sheath
was isolated naturally from the bony core without destroying the natural structure after 30 day storage at ambient conditions. Distal segments
were cut 30 mm. And then the distal part of horn was cut along with the
growth direction of the horn, and the thickness was perpendicular to
the radial direction, as shown in Fig. 1. About 2 mm external surface
and internal surface in the horns were cut away from each sample.
Samples were cut into rectangular prisms of dimensions 40 mm
5 mm 2 mm (length width thickness) with a handsaw and variety
of ne rasps. In samples' milling process, there was no overheating phenomenon in the horns.
2.2. The oxidation of horn keratin
The horn samples were washed with distilled water and 0.1 N NaCl
solution, and then immersed into ethyl ether to remove fat. After that,
the samples were washed three times with distilled water and airdried at 25 C at a relative humidity of 60 2%. Samples were treated
in an aqueous bath with a liquid-to-horn ratio of 30:1 (v/v). The hydrogen peroxide concentration was set at 20% and 30% respectively. The
bath temperature was controlled at 25 C, and the treatment time was
12 h and 48 h respectively. After the reaction, samples were immersed
in distilled water under ultrasonic vibration for 48 h, during which the
water was changed every 12 h, and then dried at 80 C under vacuum
to remove the residual hydrogen peroxide. The removal of hydrogen
peroxide was conrmed by the reduction method of potassium iodide
[20]. The oxidized horn keratins, soaked in 20% H2O2 for 12 h, 30%

Fig. 1. (A) Schematic drawing of the horn samples, showing the orientation and position
where the samples were cut. (B) The inset shows the sample orientation for tests
(not to scale).

H2O2 for 12 h and 30% H2O2 for 48 h respectively, were denoted as

OH1220, OH1230 and OH4830 respectively. In addition, the horn
keratin was denoted as OH.
2.3. Fourier transform infrared spectroscopy (FTIR)
FTIR spectra were obtained by using the attenuated total reectance
(ATR) technique of FTIR (EQUINOX55, BRUKER). An average of 32 scans
was taken from 4000 to 600 cm1 with a resolution of 4 cm1.
2.4. Thermal analysis
The thermal degradation behavior of the samples was investigated
by a thermogravimetry instrument (TG, 209F3-ASC, NETZSCH). A few
milligrams of samples were heated from 30 to 400 C at a heating rate
of 10 Cmin1, under nitrogen and air atmosphere with owing atmosphere (10 mlmin1) respectively.
The phase transition temperature of the samples was examined by a
differential scanning calorimetric instrument (DSC, 204F1, NETZSCH). A
few milligrams of samples were heated from 50 to 320 C at the heating
rate of 10 Cmin1, ushing the crucible with 100 mlmin1 nitrogen.
2.5. Tensile tests
Quasi-static tensile tests were conducted at room temperature
using a computer-controlled universal testing machine (DL-D series,
Xinzhengwei Corporation, Jiangdu, China). In order to prevent the sample
damage and slip, both the ends of the samples were pasted with square
aluminum grips (10 mm 6 mm 1 mm) by an epoxy resin adhesive,
which could transfer the load smoothly and uniformly to the two ends
of the samples. Tests were performed at a constant crosshead speed of
10 mmmin1 with a load cell of 2000 N. The force and displacement
data were automatically recorded by the built-in measurement software.
Three samples (40 mm 5 mm 2 mm) were taken from each set of the
samples for measurement, and the water content of samples was controlled at about 9%. The results were expressed as mean values standard deviation.
2.6. Scanning electron microscopy
After tensile tests, the fracture surfaces of samples were coated with
gold and observed by a scanning electron microscope (SEM, Philips
XL-30, Netherland).
2.7. In vitro degradation and SDS-PAGE analysis
Each sample (40 mm 5 mm 2 mm) was placed in a test tube containing 10 ml of phosphate-buffered saline (PBS, pH 7.4) and incubated at
37 C. The buffer solutions were replaced by fresh ones every two weeks.
After incubation, the samples were washed and dried in vacuum to constant weight. Results were expressed as percentage of weight loss (W%)
and calculated according to the equation: W = [(W0 Wt) / W0]
100%, where W0 was the weight of the dry sample at time 0 and Wt
was the weight of the dry sample at time t.
In addition, the molecular weight of the degradation products was
measured by the method of sodium dodecyl sulfate-polyacrylamide
gel electrophoresis (SDS-PAGE) [21]. At rst, in order to remove undissolved debris, the degradation solutions of samples were centrifuged
at 2500 rpm for 5 min using a microcentrifuge (SC-02, ZONKIA, Anhui,
China). Protein content in the supernatant of all samples was collected
and concentrated to be about 1 ml degradation solution respectively.
100 l of each solution was mixed with 5 l of 5 SDS loading buffer
(Bio Rad) containing 0.6 M b-mercaptoethanol (Bio-Rad). Samples
were denatured by boiling in SDS/b-mercaptoethanol solution for
5 min and then immediately placed into ice water. 30 l of these cold,
denatured solutions was loaded onto lanes of precast TrisHCl gels

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

(5% stacking gel and 12% separating gel) (Bio-Rad). Separation was performed at 80 V for approximately 3 h. After separation, gels were rinsed
with ultrapure water for 5 min before staining with Bio-Safe Coomassie
stain (G250, Bio Rad) for 10 min under boiling water bath. Destaining
was done overnight in ultrapure water with gentle rotation. Samples
were compared to a standard ladder (Benchmark Prestained Protein
Ladder, Invitrogen, Carlsbad, CA) and the gels were imaged in DIA
mode with an Image Scanner (GE, USA).
2.8. In vitro cell viability
Cell viability in the presence of horn keratin and oxidized horn keratin was assessed by an MTT assay. All keratin samples (4 mm 2 mm),
which had been soaked in PBS solution (pH 7.4) for 24 h and then
drained, were sterilized in a steam autoclave at 120 C for 30 min
prior to NIH-3T3 and human umbilical vein endothelial cell (HUVEC)
(Medical Laboratory, Jinan University, China) culture experiments. Subsequently, all the sterilized samples were placed in a 48-well culture
plate (Corning Life Sciences), and each sample had 5 duplications. The
NIH-3T3 and HUVECs were seeded at a density of 5000 cells/cm2 and
allowed to grow at 37 C atmosphere of 5% CO2. The negative control
consisted of cells without samples. After incubation of certain time,
such as 24 h, 48 h and 72 h respectively, a 3-[4,5-dimethylthiazol-2yl]-2,5-diphenyl tetrazolium bromide (MTT) solution (5 mg/ml,
Sigma) was added and incubated for further 4 h. Mitochondrial dehydrogenases of viable cells cleaved the tetrazolium ring, yielding purple
formazan crystals. Formazan crystals were then dissolved in DMSO solution (Sigma). Afterwards, 200 l of the blue solutions was transferred
to a 96-well plate. The absorbance was measured at 490 nm by a microplate reader (Bio-Rad).
2.9. In vitro hemocompatibility
The fresh rabbit's blood used in our experiments was obtained legally from the Shenzhen Testing Center of Medical Devices, China. The
analysis was performed within 12 h after blood donation. The amounts
of the samples used for statistical count were not less than three.
2.9.1. Hemolysis ratio
4 ml blood was diluted by 5 ml of 0.9% (w/v) sodium chloride solution. Each testing sample (40 mm 5 mm 2 mm) was added 10 ml
0.9% (w/v) sodium chloride solution. Additionally, 10 ml of 0.9% (w/v) sodium chloride solution and 10 ml double distilled water were prepared
respectively for antitheses. All the samples were kept at 37 C for 72 h,
and then immediately added 0.2 ml of the diluted blood. Incubation
was performed at 37 C with test tubes. After 60 min incubation, the samples were centrifuged at 750 g for 5 min. Then the supernatant was measured at 545 nm by 722 s spectrophotometer (Shanghai Sunny Hengping
Scientic Instrument Co., Ltd.). Hemolysis ratio was calculated as follows:
hemolysis

h



i
ODtest ODneg = ODpos ODneg  100%;

where ODtest, ODneg, and ODpos were the absorbance values of the test
sample, negative control (saline), and positive control (water), respectively. All the hemolysis experiments were done in triplicate.
2.9.2. Platelet adhesion and activation
Platelet adhesion experiment was carried out to evaluate the surface
thrombogenicity of the samples and to examine the interaction between blood and the materials in vitro [30]. The size of all of the samples
was 5 mm 5 mm. The rabbit whole blood was treated with anticoagulant (EDTAK2, Hunan Liuyang Medical Instrument Factory,
China). After centrifuging at 1000 rpm for 15 min, a platelet-rich plasma
(PRP) was obtained. The samples were immersed in PRP and incubated
at 37 C for 30 min. The samples were subsequently rinsed with a PBS

125

solution (pH 7.4) to remove weakly adherent platelets. The adhered

platelets were xed in 2.5% glutaraldehyde solutions at room temperature for 2 h, and then dehydrated and dried at room temperature. The
samples were then coated with gold and observed by SEM (Philips
XL-30, Netherland).
2.9.3. PTT, APTT and brinogen (Fib) assay
For the partial thromboplastin time (PTT) measurement, samples
were added to each plastic test tube. The rabbit whole blood, which contains anticoagulant (sodium citrate 1:9, Guangzhou Improve Medical
Ltd., China), was centrifuged at 2000 rpm for 15 min to obtain
platelet-poor plasma (PPP). Thereafter, 1 ml PPP was added onto the
samples (5 mm 5 mm 2 mm), which were completely immersed
and incubated at 37 C for 15 min. 100 l incubated PPP was transferred
to a test tube and 100 l PTT reagent was added to the same test tube,
followed by the addition of 0.025 M CaCl2 solution (100 l). The suspension was stirred by a magnetic stick and the coagulation time was determined at 37 C using a coagulation instrument (ACL 8000, Beckman
Coulter, Inc.).
For the activated partial thromboplastin time (APTT) test, 400 l PPP
was mixed with PBS (as a control) or the degradation solution of the
samples (200 l) and then incubated at 37 C for 15 min. After mixing,
100 l plasma mixture, 100 l APTT reagent and 100 l 0.025 M CaCl2 solution were added to the same test tube. Subsequently, the suspension
was stirred by a magnetic stick and the coagulation time was determined at 37 C using the same coagulation instrument. The Fib measurements were carried out with the same procedure of APTT, except
that 100 l Fib reagent was added.
2.10. In vivo implantation experiment
All animal procedures were performed under a protocol approved
by the Institutional Animal Care and Use Committee. To assess the
vivo biocompatibility of oxidized horn keratin biomaterials, small
autoclaved samples (0.1 g, about 8 mm 5 mm sections) were implanted subcutaneously into mice. Before implantation, the samples
were incubated in sterile PBS (pH 7.4) for 1 h. Adult female BALB/c
mice (Experimental Animal Laboratories, Guangdong, China), approximately 2025 g in weight, were implemented under general anesthesia
by intraperitoneal injections of chloral hydrate (AR, Sigma). The operative site on the back was shaved and cleansed with Betadine (Guangdong Hengjian Pharmaceutical Co., Ltd.). A 2 cm lateral skin incision
was made on the mid-portion of the back and tissue pockets were created by gross dissection laterally using blunt scissors. Sterilized samples
were implanted into the subcutaneous pocket of each mouse. Skin
closures were performed with nonabsorbable nylon sutures (Shanghai
Medical Suture Needle Co., Ltd.). At 1, 3, 6, and 10 weeks, animals
were euthanized and an incision was made on the back of each mouse
(n = 2 mice/time point, each sample). Digital photographs of samples
were taken in situ and observed grossly for inammation and capsule
formation. The entire implant site was excised, but the sample was removed because of its hard texture.
Tissue explants were xed in 10% neutral buffered formalin (NBF,
Fisher Scientic) for 48 h, and then dehydrated in increasing concentrations of ethanol (Fisher Scientic) and embedded in parafn (Fisher Scientic). Cross-sections at 5 mm thickness were cut on a microtome
(Ultramicrotome RM2235, Leica Microsystems Inc.), and mounted on
CITOGLAS glass slides, and air dried. Sections were stained with hematoxylin & eosin (H&E, Fisher Scientic) to assess the presence and thickness of brous tissue, cellular response, and vascularization using an
optical microscope (Axio Scope A1, Zeiss).
2.11. Statistical analyses
Descriptive statistics were performed on all of the experimental
data to obtain the means and the standard errors with Origin Pro7.5.

126

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

One-way analysis of variance (ANOVA) was applied to the measured

data for each experiment described above. The difference is considered
signicant when p b 0.05.
3. Results and discussion
3.1. Analysis of FTIR
During the oxidizing reaction of hydrogen peroxide, many perhydroxyl species are formed from hydrogen peroxide, such as HO
2 , which
can attack much substance, including the keratins [22,23]. The attacking
of disulde bonds by perhydroxyl species would produce disulde oxidation products from ruptured \S\S\ bonds, which are cysteic acid and
intermediate sulfoxides.
The FTIR spectra of horn keratin and oxidized horn keratins are
shown in Fig. 2. According to the curves of samples, there are three
major band regions, which should be assigned to amides I, II, III

respectively. The band in the range of 16001700 cm1 is assigned to

the amide I, which is related to the C_O stretching. While the amide
II, which is observed in 14001500 cm1, is caused by the N\H bending
and C\H stretching vibration. But the amide III, occurring in 1220
1300 cm1, results from the combination of C\N stretching and N\H
in plane bending, with some distribution from C\C stretching and
C_O bending [24,25].
In fact, the amide I is very sensitive to the secondary structure of
the proteins [26]. In order to eliminate the water absorption band
(16001650 cm1), which could interfere with the amide I, the samples
were dried in vacuum at 80 C for 2 days before the acquisition of spectra. And then, the spectra in 17101590 cm1 were baseline corrected
and smoothed with the SavitskyGolay method (9 points) [27]. At
last, the manipulated spectra were resolved by second order derivative
using the method of Marquardt, as shown in Fig. 2. The peak in 1650
1658 cm1 indicates an -helix structure, while the bands in 1640
1610 cm 1 have been assigned to -sheet [25,28]. Both -sheet and
-helix are found in the horn keratin and oxidized samples. Moreover,
the contents of two structures are different in all samples, as shown in
Table 1. With the increase in the degree of oxidation, the content of
-sheet structure increases while the content of -helical structure decreases, when the oxidized samples are compared with the horn keratin. On the other hand, the region of 12401220 cm1 corresponds to
the random coin and -sheet structure, which increases gradually
with the increase in the degree of oxidation [29,30]. These band shifts
are caused by the distinct hydrogen bonding states, which are produced
by the different protein conformations. It can indicate that the oxidized
horn keratin would change its secondary structure, for example, the oxidized horn keratin lost its ordered -helical structure to form a disordered structure, such as -sheet [31,32]. This may be due to hydrogen
peroxide which forms strong interaction with the polar side chain
groups of keratin [32]. The result is that the molecular chains become
closer. In this way, this organization can promote crystallization in sheet structure embedded in an amorphous matrix [25]. Therefore,
the amorphous matrix in oxidized samples increases gradually with
the increase in the degree of oxidation. The horn possesses a more
solid and compacted structure, but the structure of the oxidized samples
is different. Because their hydrogen bonds and disulde bonds, even
peptide bonds, are split apart, and some crystals and amorphous regions
in keratin are destroyed. In addition, band at about 1390 cm1 is related
with the C\H and O\H bending vibration, which decreases gradually
with the increase in the degree of oxidation. The disulde bond in cystine of the horn keratin is broken down to form sulfate oxides in the process of oxidation. For all the oxidized samples, a group of small peaks
with different intensities, laying in 11801022 cm1, is related to the
content of different sulfate oxides, such as cysteine acid, \SO3H [24,
33]. The intensity increases with the increase in the degree of oxidation.
It is due to the changes in some sulfur-containing groups during the oxidation. The peaks in the range of 630650 cm1 can be attributed to
the C\S band stretching vibrations [24,30]. It can be shown that the disulde bond is not completely broken down during the oxidation.
3.2. TG-DSC measurement
From the TG-DTG curves, all samples show two evident mass loss
stages, as shown in Fig. 3A. The rst stage in the temperature range of
Table 1
Characteristic of the amide bands of horn keratin and oxidized horn keratins.
Materials

-Helix
Band position (cm

Fig. 2. FTIR spectra of OH, OH1220, OH1230 and OH4830, as shown in A; the fragment
of FTIR spectra resolved into components (over range 17101590 cm1, R2 N 0.999), as
shown in B.

OH
OH1220
OH1230
OH4830

1650.8
1652.3
1653.9
1656.2

-Sheet
1

Area (%)

Band position (cm1)

Area (%)

48.74
45.80
42.42
37.66

1617.0 and 1631.2

1617.1 and 1632.3
1619.9 and 1634.0
1620.3 and 1634.3

31.69
35.35
38.15
42.10

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

127

Fig. 3. TG-DTG and DSC curves of the samples: TG-DTG (A), DSC (B). In addition, a, b, c and d for OH, OH1220, OH1230 and OH4830 respectively.

30150 C generally corresponds to the evaporation of moisture, and

the second stage is assigned to the thermal degradation of samples,
which occurs in the temperature range of 220400 C.
With the increase in the degree of oxidation, the mass loss of oxidized samples in the temperature range of 30150 C decreases, namely
about 7.93%, 7.61% and 6.58% for OH1220, OH1230 and OH4830 respectively. But the mass loss of horn keratin is about 6.82% in the temperature range of 30150 C. From the DTG curves, the temperature in
thermal degradation of horn keratin is higher than that of oxidized samples and the temperature peak is broader. However, as for oxidized
samples, the order of temperature in thermal degradation is OH48
30 N OH1230 N OH1220. This suggests that the oxidized samples
have lower thermal stability compared to the horn keratin; in addition,

the thermal stability of oxidized samples would increase with the increase in degree of oxidation.
The DSC curves of samples are shown in Fig. 3B. There are three endothermic peaks in the DSC curves, and they roughly correspond to two
evident mass losses. For horn keratin, the well-known endothermic
peak at about 92 C resulted from water evaporation and glass transition. Due to the complex structure of horn keratin, the glass transition
often occurs in a temperature range rather than at a xed temperature
[34,35]. Moreover, this temperature range is usually overlapped with
the peak of water evaporation in a DSC curve [36]. The endothermic
peak at 236 C is ascribed to the denaturation of the -keratin crystallites and the area under the curve can be used to measure the -helix
content; and the endothermic peak at 312 C corresponds to the

128

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

destruction of crosslinks, such as disulde bonds, hydrogen bonds, and

salt links [37,38]. However, the peaks of oxidized samples are different
from those of horn keratin, suggesting that the microstructure of the oxidized samples changed during the process of oxidation.
It is well known that horn keratin is made up of amorphous matrix and crystalline regions [10]. The former provides weakly bound
water sites, but the latter provides strongly bound structural water sites
[3]. From FTIR, the amount of hydrophilic groups, such as \COOH
and\SO3H, and the amorphous matrix in oxidized samples increases
during the process of oxidation, which can present more weakly bound
water sites as well as increase the afnity to water in the oxidized samples
when compared with the horn keratin. Therefore, more hydrophilic
groups absorb water and more moisture-bonded structures are formed
in oxidized samples. This corresponds to the rst mass loss of samples
in TG curves. Moreover, the temperature of water evaporation in oxidized
samples increases from 84 C to 107 C. But the temperature at about 92
C in horn keratin is caused by the presence of strongly bound water,
mainly provided by crystalline phase. It indicates that more amounts of
hydrophilic groups provide more strongly bound water than that in
crystalline phase.
The results from DSC curves also show that as the increase in the degree of oxidation, the denaturation temperature (246 C, 245 C, 242 C
respectively) for -keratin crystallites is higher than that in horn keratin; however, the temperature (295 C, 296 C, 298 C respectively) for
destruction of crosslinks is lower than that in horn keratin. The lower
denaturation temperature of horn keratin may be resulted from the
lower amount of crystalline -sheet structure. In -sheet structure, intermolecular interaction between the protein chains is stronger than
that in the keratin bers [39]. However, with increase in the degree of
oxidation, -helical structure decreases evidently, which makes the
change of crystal and the reduction of the crystallinity, resulting in the
decrease of the denaturation temperature. On the other hand, the crystalline denaturation peak is broader in horn keratin, which can reect a
distribution of crystal sizes in horn keratin [40]. In addition, the better
thermal-stability of horn keratin is a result of the cross-linking between
the macromolecules by more disulde bonds and hydrogen bonds. After
oxidation, the decrease of disulde bonds weakens the crosslinks, or
even the crosslinks are broken down, diminishing the stability to heat
and resulting in the decrease of the destruction temperatures of the
crosslinks. However, with the increase in the degree of oxidation, only
the crosslinks that are not easily to be destroyed are left. Furthermore,
the increase of amorphous matrix will produce more hydrogen bonds
[34]. Therefore, the destruction temperature of crosslinks in oxidized
horn keratin increases gradually with the increase in the degree of oxidation, which is in accordance with the performance of the samples in
TG-DTG curves.
3.3. The effects of oxidation on the mechanical properties of horn keratin
During the oxidizing reaction of hydrogen peroxide, the disulde
bond and hydrogen bond, or even peptide bond, are attacked and partly
broken down, which has signicant effects on the mechanical properties of horn keratin [41]. In previous studies, the tensile strength,
Young's modulus and fracture strain of untreated samples with water
content of 9% were found to be 117.58 3.13 (MPa), 1.553 0.031
(GPa) and 24.60 3.60 (%) respectively [13]. In comparison, the tensile
strength and Young's modulus of the oxidized horn keratin are moderately degraded, and the failure strain increases with the increase in the
degree of oxidation (p b 0.05), as shown in Table 2. Despite that, the oxidized samples could be fully recovered to its original shape and dimensions after the tensile test.
The cross-linkages in keratin are formed by disulde bonds, together
with the van der Waals forces, hydrogen bonds and ionic interaction.
The cross-linkages contribute to mechanical properties as well as structural stability, making the horn keratin have better stiffness and
strength [13]. After oxidation, the disulde bonds, van der Waals forces,

Table 2
The values of mechanical characteristics in horn keratin and oxidized horn keratins.
Samples

Tensile strength (MPa)

Young's modulus (GPa)

Fracture strain (%)

OH
OH1220
OH1230
OH4830

117.581
100.011
85.758
72.680

1.553
1.137
0.869
0.732

24.602
37.566
56.180
70.529

3.129
2.341
5.386
2.063

0.031
0.029
0.021
0.022

3.599
4.113
4.252
6.371

n = 3 for each set of samples. Values are means s.e.m. and p b 0.05 by comparisons at
all samples.

hydrogen bonds and ionic interaction, or even peptide bonds, are broken somewhat, resulting in less coherent interactions within the protein
structure [42]. On the other hand, the crystalline regions with -helical
structure are responsible for the strength of horn keratin, but the amorphous regions, which have relatively fewer bonds between the polymer
chains and random distribution of the chains, provide the horn keratin
with elasticity and exibility [43]. From the results of FTIR, the
-helical structure decreases and amorphous matrix increases after oxidation. As a result, this would cause larger matrix region with freedom
of movement and reduce the stability of the matrix, resulting in the
decrease of the tensile strength and Young's modulus and the improvement of failure strain of oxidized horn keratin. However, high concentration of hydrogen peroxide might contribute to the production of
excessive amounts of perhydroxyl species, which can react with more
substances. If the time of oxidation is longer, it would let perhydroxyl
species have more time to attack the proteins [41]. This can cause further breakage of bonds, even polypeptide chains. Therefore, as the degree of oxidation increases, there is a signicant reduction in tensile
strength and Young's modulus in oxidized horn keratin, but signicant
increase in toughness. It indicates that the mechanical properties of
horn keratin are largely affected by the concentration of hydrogen peroxide and the time of oxidation. However, the tensile strength and fracture strain of oxidized horn keratin are stronger than those of other
synthetic materials, such as polycarbonate (67 MPa and 15%) [44] and
polylactide (65 MPa and 9%) [45]; even the tensile strength of OH12
20 can rival that of berglass (110 MPa) [46]. Therefore, with appropriate oxidation treatment, the oxidized horn keratin can maintain its
mechanical property not to change too much.
3.4. Microstructure of fracture surfaces
The horn keratin is a hierarchical material and has laminate structure [8]. Related to the fracture surfaces of oxidized samples, the SEM
images of samples reveal different failure phenomena, as seen in
Fig. 4. It is clearly shown that the fracture surface in buffalo horn is relatively smooth, neat and wavy, indicating that the buffalo horn has a
dense laminate structure. But the bers are pulled out and the lamellas
are partially torn in OH1220. Moreover, the fracture surfaces of OH12
30 and OH4830 show an extremely ductile fracture mode, evidenced
by a very deep, convoluted cup-and-cone type fracture. At the same
time, the samples had fully recovered to their original shape and dimension after a certain time. The larger failure strain also indicates that oxidized horn keratin is more resilient than horn keratin. It may be due to
the more compliant matrix that can yield and ow more readily with
the increase in the degree of oxidation, which is corresponding to the
results of the mechanical properties of the samples.
3.5. In vitro hydrolytic stability and SDS-PAGE analysis
Numerous disulde bonds permanently bind the peptide chains,
which contributes to the insolubility or low insolubility of keratin in
water. The degradation behavior of horn keratin after oxidation was investigated using an in vitro degradation experiment, which would provide a good understanding in the hydrolytic stability of oxidized horn
keratin. From the weight loss ratiotime curve in Fig. 5, the degradation
rate of OH is extremely slow, nearly no degradation during 10 weeks.

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

129

Fig. 4. The fracture surfaces of the samples under tensile test. A for OH, B for OH1220, C for OH1230 and D for OH4830.

However, with the disulde bonds being oxidized to break down, the
hydrolytic stability of oxidized horn keratin decreases gradually.
Compared with the horn keratin, there is a signicant difference in the
degradation rate of oxidized samples (p b 0.05). The degradation rate
of oxidized horn keratins gradually accelerates with the increase in
degree of oxidation. For example, after 16 weeks, the degradation rate
of OH4830 is up to about 90%, nearly completely degraded; but for
OH1220, the weight loss ratiotime curve is substantially linear,
which indicates that the degradation behavior of OH1220 is relatively
stable. However, there is a sudden sharp rise in weight loss ratiotime
curve of OH1230 and OH4830. As expected, with more broken disulde
bonds, the degradation behavior of the oxidized horn keratin is signicantly enhanced, resulting in less hydrolytic stability.
In SDS-PAGE method, the mobility of protein depends on its relative
molecular mass, regardless of the electric charge and molecular shapes.
Therefore, after the qualitative analysis of protein fractions in degradation
products by SDS-PAGE, the protein bands from 3 separate degradation
products of oxidized horn keratins reveal similar patterns, including
high molecular mass bands (N40 kDa) and low molecular mass bands
(b 25 kDa), as shown in Fig. 6. The protein bands at about 4060 kDa

Fig. 5. The weight loss ratiotime curves of the samples. Each value is expressed as
mean standard deviation.

are attributed to monomeric keratin subunits that are mainly low-sulfur

content of -helical keratins; however, the lesser low molecular mass
bands at about 1520 kDa are attributed to high-sulfur content of matrix
[47]. In addition, the protein bands at about 60160 kDa may be assigned
to obligate keratin heterodimers. Due to the proteinprotein interaction
or cross-linkings, the higher molecular mass bands (N 160 kDa) may correspond to stable tetramers or larger multimers of keratins [48]. The molecular mass of degradation products in three oxidized horn keratins is
mainly more than 20 kDa, with less of low molecular mass; moreover,
for OH1220 and OH1230, protein bands are mainly focused on the region of high molecular mass, even higher protein bands (N160 kDa).
However, compared with OH1220 and OH1230, the molecular mass
of degradation product of OH4830 is decreased to some extent. This
may be depended on the extent of damage to the structure of horn keratin
by oxidation. For example, the destruction of disulde bonds or even
peptide bonds would make the protein chains interrupt randomly to be
protein fragments, which decreases the molecular mass of protein. With
the increase in degree of oxidation, single protein will be cut into more

Fig. 6. SDS-PAGE of degradation solution of oxidized horn keratins.

130

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

Table 3
Hemolysis ratio and PTT of horn keratin and oxidized horn keratins. The hemolysis ratio of
the positive control (water) and negative control (saline) was 1 and 0, respectively.
Samples

Hemolysis ratio (%)

PTT (s)

Original plasma
Positive control
OH
OH1220
OH1230
OH4830

1.80
1.54
1.99
1.49

63.23
30.57
63.55
60.73
61.13
59.33

0.35
0.23
0.16
0.77

1.17
0.93
1.21
1.86
3.49
1.76

protein fragments by more perhydroxyl species, resulting that the protein

fragments have lower molecular mass. As a result, the protein fragments
are released during the degradation, reected in the SDS-PAGE. However,
the mechanism of degradation in the oxidized horn keratin needs to be
proved by more experiments.
3.6. In vitro hemocompatibility
3.6.1. Hemolysis ratio
Hemolysis ratio is an important factor to evaluate the blood compatibility of a biomaterial. When the red blood cells swell to the critical
bulk, which would make the cell membranes break up, hemolysis is
formed. In this way, the adenosine diphosphate is released from the
broken red blood cells, which can intensify the assembly of platelets.
As a result, the formation of clot and thrombus is accelerated [49]. If
the hemolysis ratio is lower, the blood compatibility is better with less
broken red blood cells. The hemolysis ratios of all samples in Table 3
are all well within 5%. As a novel material contacting with the blood, if
its hemolysis ratio is less than the accepted threshold value of 5%, it
has a good hemocompatibility [50]. It directly demonstrates that the oxidized horn keratin can be used as biomaterials without causing the red
blood cell to have change in deformability and fracture to form any
hemolysis.
3.6.2. Platelet adhesion and activation
Platelet adhesion and activation are known as the main intuitive indicators and often used to assess the hemocompatability of materials
[51]. If platelets are spreading and aggregating on the surfaces of materials, the platelets are activated. It is a major mechanism for the formation of thrombosis. On the surface of the tested samples, some adherent
platelets are in a moderate degree aggregation, but most adherent

Fig. 8. APTT and Fib of the original plasma and the plasma contacted with different concentrations of degradation product of the oxidized keratins.

platelets still remain individual and spherical, separated without pseudopodium, as shown in Fig. 7. In addition, the adherent platelets are
not found to further induce a large number of platelet to aggregate.
The process of blood coagulation is initiated when platelets are aggregating with the formation of a brin network. After that, a thrombus is
subsequently formed [52]. However, if the surface of material is passivated by a thin layer of platelets, without activation, it will have better
hemocompatibility [53]. As there are no platelet aggregation and activation on the surfaces of oxidized horn keratins, it directly demonstrates
that the oxidized horn keratins would not activate blood clotting system
to form thrombus.
3.6.3. Fibrinogen activation
Fibrinogen is a serum protein and plays a dominant role in the formation of thrombus [54]. The Fib levels of samples fall within normal
level, without showing signicant difference with each other (p N
0.05), but have a signicant difference to positive control (p b 0.05),
as shown in Fig. 8. Generally, the increase of Fib would enhance the
blood coagulation to increase the thrombus formation. This may be
due to the conformational changes of brinogen, which would make

Fig. 7. Morphology of adherent platelets on the surfaces of the samples. A for OH, B for OH1220, C for OH1230 and D for OH4830.

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

131

Fig. 9. The results of MTT assay of the samples, A for 3T3 cell and B for HUVECs. Each value is expressed as mean standard deviation.

brinogen to combine with the GPIIb/IIIa integrin receptor on platelet

membrane and further trigger the platelets to aggregate [55]. The
normal Fib levels show that the degradation products of oxidized horn
keratin would not activate the platelets to trigger thrombus.
3.6.4. PTT and APTT
Coagulation cascade system has intrinsic and extrinsic pathways,
which both converge at a common point. When the factor X is activated

to Xa, prothrombin is sequentially activated to convert to thrombin,

which would trigger and accelerate the formation of brin from brinogen [56]. The intrinsic pathway is initiated when the material is
contacted with the blood, which would sequentially activate the
clotting process to lead to thrombosis. In addition, it is well known
that PTT and APTT are used to detect the abnormalities of factors in
intrinsic pathway, such as factors I, II, V, VIII, IX, X, XI, and XII, and brinogen [57]. As shown in Table 3, the PTTs of samples do not show

Fig. 10. The peri-implant tissue was obtained at 1 week to 10 weeks. Gross examination of the implant area did not show any observable inammation or capsule formation with time
elapsing. AD for OH, EH for OH1220, IL for OH1230 and MP for OH4830 respectively.

132

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

Fig. 11. The H&E staining of the peri-implant tissue. AD for OH; EH for OH1220; IL for OH1230 and MP for OH4830 respectively.

signicant difference with each other (p N 0.05), but have a signicant

difference to positive control (p b 0.05). PTT, without adding activators,
can reveal whether there is activation between plasma and the material.
Shortening of the PTT will increase the risk of thromboembolism. From

the results of PTT falling within the normal level, the horn keratin and
the oxidized horn keratin would not activate the platelets. On the
other hand, compared with the positive control and the original plasma,
the APTT variation falls within the normal level (p N 0.05) for all the

Fig. 12. The high-magnication micrograph of H&E staining. AD for OH; EH for OH1220; IL for OH1230 and MP for OH4830 respectively.

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

tested concentrations of degradation solution, as seen in Fig. 8. It indicates that the degradation solution (01 mg/ml) does not effectively
benet to the activation of intrinsic blood coagulation system.
However, the keratin biomaterials, which were extracted from wool,
hair, etc., have been demonstrated to be as an efcient hemostatic agent
in several animal models [58]. In this way, the extracted keratin does
not have three-dimensional structure, and the main component of its
degradation is a low molecular weight of polypeptide or protein. But it
is different from the oxidized horn keratin of this article, as can be
seen from the results of FTIR, DSC, mechanical properties and SDSPAGE, which may cause the oxidized horn keratin not to play the role
of procoagulant. From the results on hemocompatibility analysis, it
can obviously show that horn keratin and oxidized horn keratins
would not interfere with the normal functioning of platelets, without
signicant inuence on the coagulation system. It directly demonstrates
that the oxidized horn keratins have better security because they meet
with the basic requirements of hemocompatibility of biomaterials.
3.7. Cell viability
The cells observed by optical microscopy grow well with normal cell
morphology both in control group and the test group; moreover, there
are discrete particles within the cytoplasm, without cytolysis. The MTT
assay is usually used to evaluate the cytotoxicity of material by quantifying relative cell numbers [19]. As shown in Fig. 9, the horn keratin and
oxidized horn keratin have no signicant effects on the viability of 3T3
and HUVECs when exposed to cells for 24 h, 48 h and 72 h respectively.
Compared to serum-containing media, there is no signicant evidence
of cytotoxicity and cell viability of samples is statistically equivalent. It
is demonstrated that the horn keratin and oxidized horn keratin have
non-cytotoxicity in vitro and compliance with the requirements of
biomaterials.
3.8. In vivo tissue response
After implantation, the material would be seen as a foreign body and
attacked by the immune system of the recipient. In the absence of other
factors, if the material is toxic, it causes death of the surrounding tissue;
if the material is nontoxic, the reaction between the tissue and implants
is primarily aseptic inammation and brous capsule [59]. Early tissue
response is mild or moderate acute aseptic inammation, such as
edema, hyperemia, and neutrophil inltration, which is caused by implant irritation. After two weeks, the acute inammatory would change
to chronic inammation, including macrophages, lymphocytes and broblast proliferation. Organism eliminates foreign body through
phagocytosis and enzymatic digestion, or by brous capsule wrapping
to insulate implants [60,61].
During the experiments, the activities and diet of experimental mice
were normal, without accidental death. In early stage of implantation,
the dissected inner side of the skin shows acute inammation, such as
hyperemia around the implants; and then the inammatory response
changes to chronic inammation, as shown in Fig. 10. At later time
points, hyperemia in the subcutaneous tissue decreases, particularly
around the implants, which can reect that the inammation is diminished. In addition, the implants are isolated by brous capsule, which
could make the implants not be eliminated by the cellular immune system. The thickness of brous capsule may reect the histocompatibility
between implants and tissue [48,62]. If the capsule is thicker, the foreign
body reaction would be heavier, resulting worse histocompatibility; and
vice versa. However, the results do not exhibit thicker brous capsule at
all-time points. So it can demonstrate that there are no adverse reactions between the implants and tissue, without obvious inammation
or brosis at late time points.
From the results of histological section in the vicinity of samples, it
can conrm that host cell migrates, such as inammatory cells, endothelial cells, macrophages and broblasts, as shown in Figs. 11 and 12.

133

In early stage of the implantation, the peri-implant tissue response

does not appear toxic reaction, such as cell lysis and destruction, and appears to be predominantly neutrophils and a small amount of lymphocytes. Neutrophils are the rst responders for the acute inammation.
After that, macrophages, endothelial cells and broblasts are coming
[63]. After 3 weeks, neutrophils have decreased, but there is an increase
in the number of lymphocytes, macrophages and activated broblasts.
However, the total number of inammatory cells on the periphery of
implants signicantly reduces, revealing the chronic inammation.
Later, the peri-implant tissue response is dominated by mature broblasts with a small number of lymphocytes. Inammatory cell accumulation reaches the maximum number within 13 weeks and then
gradually goes down, which may be due to the short life-time of neutrophils and macrophages. It indicates that the inammation decreases
during the time. On the other hand, foreign body giant cells also exist
in all time points, which can reveal that the oxidized horn keratins are
highly compatible with tissue.
From the above results of assessments, the insertions of horn keratin
and oxidized horn keratins subcutaneously implanted into mice do not
cause a substantial inammatory reaction. All responses around the implants are limited to mild foreign body reactions. Thus it demonstrates
that horn keratin and oxidized horn keratins have good biocompatibility and can act as an implanted biomaterial for clinical applications.
4. Conclusions
Overall, our study can demonstrate some general properties of oxidized horn keratin and support its use as a biomaterial. The FTIR reveals
that the disulde bonds, or even partial peptide chain of horn keratin,
are broken down during oxidation. With the change in microstructure,
the oxidized horn keratin has lower thermal stability and hydrolytic stability in comparison with horn keratin. The mechanical properties of
oxidized horn keratins are poorer than those of horn keratin, but the
oxidized horn keratins still have disulde bonds to form a threedimensional structure, which benets for their mechanical properties.
However, the fracture toughness of oxidized horn keratin increases
with the increase in the degree of oxidation. After oxidation, the oxidized horn keratins have lower cytotoxicity and lower hemolysis ratio.
Moreover, when the oxidized horn keratins, as well as different concentrations of degradation products of oxidized horn keratins, are directly
in contact with platelet-rich plasma, platelets are not activated. It suggests that the oxidized horn keratins have good hemocompatibility,
without triggering blood thrombosis. The implantation experiment
in vivo also demonstrates that the oxidized horn keratins are compatible with the tissue, because there are minimal brous capsule
and less of inltration of host cells, without causing serious inammation. Therefore, the oxidized horn keratin may offer a potential
application of implanted biomaterial that is directly in contact
with blood and tissue.
Acknowledgments
This work has been nancially supported by the National Natural
Science Foundation of China (Grant No. 20976068/B060805) and
the Science and Technology Program of Guangdong Province (No.
268017). In addition, the authors would like to thank Shenzhen
Testing Center of Medical Devices, China. They would also like to
thank Yuan Tian, coming from Biomedical Engineering of Jinan
University.
References
[1] A. Giroud, C. Leblond, Ann. N. Y. Acad. Sci. 53 (1951) 613626.
[2] Y. Noishiki, H. Ito, T. Miyamoto, H. Inagaki, Kobunshi Ronbunshu 39 (1982)
221227.
[3] T.Z. Rizvi, M.A. Khan, Int. J. Biol. Macromol. 42 (2008) 292297.
[4] M. Zoccola, A. Aluigi, C. Tonin, J. Mol. Struct. 938 (2009) 3540.

134
[5]
[6]
[7]
[8]
[9]
[10]
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
[22]
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
[33]

Q. Zhang et al. / Materials Science and Engineering C 47 (2015) 123134

J. Douglas, C. Mittal, J. Thomason, J. Jofriet, J. Exp. Biol. 199 (1996) 18291836.
A. Franck, G. Cocquyt, P. Simoens, N.D. Belie, Biosyst. Eng. 93 (2006) 459467.
M. Feughelman, M. Robinson, Text. Res. J. 41 (1971) 469474.
L. Tombolato, E.E. Novitskaya, P.Y. Chen, F.A. Sheppard, J. McKittrick, Acta Biomater.
6 (2010) 319330.
P.Y. Chen, J. McKittrick, M.A. Meyers, Prog. Mater. Sci. 57 (2012) 14921704.
J.F. Vincent, J.D. Currey, The Mechanical Properties of Biological Materials,
Cambridge University Press, Britain, 1980.
M.W. Trim, M.F. Horstemeyer, H. Rhee, H. El Kadiri, L.N. Williams, J. Liao, K.B.
Walters, J. McKittrick, S.J. Park, Acta Biomater. 7 (2011) 12281240.
H. Zahn, J. Fhles, M. Nlenhaus, A. Schwan, M. Spel, Ind. Eng. Chem. Res. 19 (1980)
496501.
Q.B. Zhang, C. Li, Y.T. Pan, G.H. Shan, P. Cao, J. He, Z.S. Lin, N.J. Ao, Y.X. Huang, Mater.
Sci. Eng. C 33 (2013) 50365043.
J.H. Buchanan, Biochem. J. 167 (1977) 489.
C. Earland, C. Knight, Biochim. Biophys. Acta 22 (1956) 405411.
J. Maclaren, Aust. J. Chem. 15 (1962) 824831.
I. O'donnell, E. Thompson, Aust. J. Biol. Sci. 17 (1964) 973978.
J. Li, Y. Li, Y. Zhang, X. Liu, Z. Zhao, J. Zhang, Y. Han, D. Zhou, Food Chem. Toxicol. 57
(2013) 338345.
A. Vasconcelos, A.C. Paulo, Curr. Drug Targets 14 (2013) 612619.
S. Merouani, O. Hamdaoui, F. Saoudi, M. Chiha, J. Hazard. Mater. 178 (2010)
10071014.
U.K. Laemmli, Nature 227 (1970) 680685.
M. Nagarajan, S. Benjakul, T. Prodpran, P. Songtipya, P. Nuthong, Food Chem. 138
(2013) 11011108.
X. Wang, G. Cao, W. Xu, J. Appl. Polym. Sci. 112 (2009) 19591966.
J. Zhang, Y. Li, J. Li, Z. Zhao, X. Liu, Z. Li, Y. Han, J. Hu, A. Chen, Powder Technol. 246
(2013) 356362.
A. Aluigi, M. Zoccola, C. Vineis, C. Tonin, F. Ferrero, M. Canetti, Int. J. Biol. Macromol.
41 (2007) 266273.
W.K. Surewicz, H.H. Mantsch, D. Chapman, Biochemistry 32 (1993) 389394.
D.W. Marquardt, J. Soc. Ind. Appl. Math. 11 (1963) 431441.
J. Kong, S. Yu, Acta Biochim. Biophys. Sin. 39 (2007) 549559.
D.B. DeOliveira, W.R. Trumble, H.K. Sarkar, B.R. Singh, Appl. Spectrosc. 48 (1994)
14321441.
H. Edwards, D. Hunt, M. Sibley, Spectrochim. Acta A 54 (1998) 745757.
S.Y. Lin, K.J. Duan, T.C. Lin, Methods Find. Exp. Clin. Pharmacol. 18 (1996) 175181.
A. Vasconcelos, G. Freddi, A.C. Paulo, Biomacromolecules 9 (2008) 12991305.
P. Erra, N. Gomez, L. Dolcet, M. Julia, D. Lewis, J. Willoughby, Text. Res. J. 67 (1997)
397401.

[34] X. Liu, S. Gu, W. Xu, J. Therm. Anal. Calorim. 111 (2012) 259266.
[35] P.R. Brady, Finishing and Wool Fabric Properties: A Guide to the Theory and Practice
of Finishing Woven Wool Fabrics, CSIRO Wool Technology, Geelong, 1997.
[36] M. Marti, R. Ramirez, A. Manich, L. Coderch, J. Parra, J. Appl. Polym. Sci. 104 (2007)
545551.
[37] R.M.D. Gama, T.S. Balogh, S. Frana, T.C.S. Dias, V. Bedin, A.R. Baby, J.D.R. Matos,
M.V.R. Velasco, J. Therm. Anal. Calorim. 106 (2011) 399405.
[38] C. Tonin, A. Aluigi, M. Bianchetto Songia, C. D'arrigo, M. Mormino, C. Vineis, J. Therm.
Anal. Calorim. 77 (2004) 987996.
[39] I.C. Um, H. Kweon, Y.H. Park, S. Hudson, Int. J. Biol. Macromol. 29 (2001) 9197.
[40] J.R. Barone, W.F. Schmidt, C.F.E. Liebner, J. Appl. Polym. Sci. 97 (2005) 16441651.
[41] X. Wang, X. Shen, W. Xu, Appl. Surf. Sci. 258 (2012) 1001210016.
[42] Q. Li, T. Lin, X. Wang, J. Text. Inst. 103 (2012) 662668.
[43] Q. Li, C.J. Hurren, H. Yu, C. Ding, X. Wang, Text. Res. J. 82 (2011) 195202.
[44] S. Hashemi, J. Mater. Sci. 35 (2000) 58515856.
[45] S.L. Yang, Z.H. Wu, W. Yang, M.B. Yang, Polym. Test. 27 (2008) 957963.
[46] W.D. Callister, Materials Science and Engineering: An Introduction, John Wiley and
Sons, New York, 2007.
[47] J.E. Plowman, J. Chromatogr. B 787 (2003) 6376.
[48] R.C.D. Guzman, M.R. Merrill, J.R. Richter, R.I. Hamzi, O.K.G. Roberts, M.E.V. Dyke, Biomaterials 32 (2011) 82058217.
[49] J.P. Singhal, A.R. Ray, Biomaterials 23 (2002) 11391145.
[50] L. Li, M. Tu, S. Mou, C. Zhou, Biomaterials 22 (2001) 25952599.
[51] Y.J. Kim, I.K. Kang, M.W. Huh, S.C. Yoon, Biomaterials 21 (2000) 121130.
[52] C. Mao, Y. Qiu, H. Sang, H. Mei, A. Zhu, J. Shen, S. Lin, Adv. Colloid Interface Sci. 110
(2004) 517.
[53] O. Moen, E. Fosse, V. Brockmeier, C. Andersson, T.E. Mollnes, K. Hgsen, P. Venge,
Ann. Thorac. Surg. 60 (1995) 13171323.
[54] W.C. Lin, T.Y. Liu, M.C. Yang, Biomaterials 25 (2004) 19471957.
[55] G. Li, P. Yang, W. Qin, M.F. Maitz, S. Zhou, N. Huang, Biomaterials 32 (2011)
46914703.
[56] A. Sasidharan, L.S. Panchakarla, A.R. Sadanandan, A. Ashokan, P. Chandran, C.M.
Girish, D. Menon, S.V. Nair, C.N.R. Rao, M. Koyakutty, Small 8 (2012) 12511263.
[57] J.B. Miale, Acad. Med. 37 (1962) 1147.
[58] M.B. Rahmany, R.R. Hantgan, M. Van Dyke, Biomaterials 34 (2013) 24922500.
[59] E.J. Kaminski, R.J. Oglesby, N.K. Wood, J. Sandrik, J. Biomed. Mater. Res. 2 (1968)
8188.
[60] J. Cohen, J. Bone Joint Surg. Am. 41 (1959) 152166.
[61] J.M. Anderson, A. Rodriguez, D.T. Chang, Semin. Immunol. 20 (2008) 86100.
[62] J.M. Morais, F. Papadimitrakopoulos, D.J. Burgess, AAPS J. 12 (2010) 188196.
[63] T. Velnar, T. Bailey, V. Smrkolj, J. Int. Med. Res. 37 (2009) 15281542.